What are the arguments in favor of delay/force in eval.c?

[Rob Browning]

Does anyone know what the arguments are, if any, for implementing
delay and force directly in eval.c rather than more generically, at
the Scheme level, perhaps in boot-9.scm via define-record, lambda,
etc.?

  (define-record promise
    ...)

  (define-syntax (delay exp) ... (make-promise ... (lambda () (exp) ...) ...))

  (define-syntax (force promise) ... ((promise-get-thunk promise)) ...)

Is the primary argument efficiency?

I ask because I was thinking about SRFI-45 (perhaps in prelude to
SRFI-40) and was trying to determine what Guile specific constraints
might apply to an implementation.

   SRFI-45: Primitives for Expressing Iterative Lazy Algorithms
   SRFI-40: A Library of Streams

If these were added to Guile, then we might want the SRFI-40 delay and
force to just replace our existing implementations.

Thanks
-- 
Rob Browning
rlb @defaultvalue.org and @debian.org; previously @cs.utexas.edu
GPG starting 2002-11-03 = 14DD 432F AE39 534D B592  F9A0 25C8 D377 8C7E 73A4